scholarly journals Label GM-PHD Filter Based on Threshold Separation Clustering

Sensors ◽  
2021 ◽  
Vol 22 (1) ◽  
pp. 70
Author(s):  
Kuiwu Wang ◽  
Qin Zhang ◽  
Xiaolong Hu
Keyword(s):  
Computational Efficiency ◽  
Tracking Error ◽  
Gaussian Mixture ◽  
Tracking Performance ◽  
Gaussian Component ◽  
Probability Hypothesis Density ◽  
Tracking Process ◽  
Multi Target Tracking ◽  
Finite Set ◽  
Environment Simulation

Gaussian mixture probability hypothesis density (GM-PHD) filtering based on random finite set (RFS) is an effective method to deal with multi-target tracking (MTT). However, the traditional GM-PHD filter cannot form a continuous track in the tracking process, and it is easy to produce a large number of redundant invalid likelihood functions in a dense clutter environment, which reduces the computational efficiency and affects the update result of target probability hypothesis density, resulting in excessive tracking error. Therefore, based on the GM-PHD filter framework, the target state space is extended to a higher dimension. By adding a label set, each Gaussian component is assigned a label, and the label is merged in the pruning and merging step to increase the merging threshold to reduce the Gaussian component generated by dense clutter update, which reduces the computation in the next prediction and update. After pruning and merging, the Gaussian components are further clustered and optimized by threshold separation clustering, thus as to improve the tracking performance of the filter and finally realizing the accurate formation of multi-target tracks in a dense clutter environment. Simulation results show that the proposed algorithm can form a continuous and reliable track in dense clutter environment and has good tracking performance and computational efficiency.

scholarly journals Multisensor Multi-Target Tracking Based on GM-PHD Using Out-Of-Sequence Measurements

Sensors ◽  
2019 ◽  
Vol 19 (19) ◽  
pp. 4315 ◽  
Author(s):  
Meiqin Liu ◽  
Tianyi Huai ◽  
Ronghao Zheng ◽  
Senlin Zhang
Keyword(s):  
Target Tracking ◽  
Tracking System ◽  
Gaussian Mixture ◽  
Tracking Performance ◽  
Probability Hypothesis Density ◽  
Single Target ◽  
Multi Target Tracking ◽  
Finite Set ◽  
Single Target Tracking ◽  
Better Than

In this paper, we study the issue of out-of-sequence measurement (OOSM) in a multi-target scenario to improve tracking performance. The OOSM is very common in tracking systems, and it would result in performance degradation if we used it inappropriately. Thus, OOSM should be fully utilized as far as possible. To improve the performance of the tracking system and use OOSM sufficiently, firstly, the problem of OOSM is formulated. Then the classical B1 algorithm for OOSM problem of single target tracking is given. Next, the random finite set (RFS)-based Gaussian mixture probability hypothesis density (GM-PHD) is introduced. Consequently, we derived the equation for re-updating of posterior intensity with OOSM. Implementation of GM-PHD using OOSM is also given. Finally, several simulations are given, and results show that tracking performance of GM-PHD using OOSM is better than GM-PHD using in-sequence measurement (ISM), which can strongly demonstrate the effectiveness of our proposed algorithm.

scholarly journals Time-Matching Random Finite Set-Based Filter for Radar Multi-Target Tracking

Sensors ◽  
2018 ◽  
Vol 18 (12) ◽  
pp. 4416 ◽  
Author(s):  
Defu Jiang ◽  
Ming Liu ◽  
Yiyue Gao ◽  
Yang Gao ◽  
Wei Fu ◽  
...  
Keyword(s):  
Target Tracking ◽  
Gaussian Mixture ◽  
Bayesian Filtering ◽  
Probability Hypothesis Density ◽  
Random Finite Set ◽  
Bayesian Formulation ◽  
Multi Target Tracking ◽  
Finite Set ◽  
Density Filter ◽  
Bernoulli Filter

The random finite set (RFS) approach provides an elegant Bayesian formulation of the multi-target tracking (MTT) problem without the requirement of explicit data association. In order to improve the performance of the RFS-based filter in radar MTT applications, this paper proposes a time-matching Bayesian filtering framework to deal with the problem caused by the diversity of target sampling times. Based on this framework, we develop a time-matching joint generalized labeled multi-Bernoulli filter and a time-matching probability hypothesis density filter. Simulations are performed by their Gaussian mixture implementations. The results show that the proposed approach can improve the accuracy of target state estimation, as well as the robustness.

scholarly journals Improved Gaussian Mixture Probability Hypothesis Density for Tracking Closely Spaced Targets

2017 ◽  
Vol 63 (3) ◽  
pp. 247-254 ◽  
Author(s):  
Huanqing Zhang ◽  
Hongwei Ge ◽  
Jinlong Yang
Keyword(s):  
State Estimation ◽  
Analytic Solution ◽  
Gaussian Mixture ◽  
Multiple Target ◽  
Probability Hypothesis Density ◽  
Random Finite Set ◽  
Multi Target Tracking ◽  
Finite Set ◽  
Estimation Precision ◽  
Simulation Results

AbstractProbability hypothesis density (PHD) filter is a suboptimal Bayesian multi-target filter based on random finite set. The Gaussian mixture PHD filter is an analytic solution to the PHD filter for linear Gaussian multi-target models. However, when targets move near each other, the GM-PHD filter cannot correctly estimate the number of targets and their states. To solve the problem, a novel reweighting scheme for closely spaced targets is proposed under the framework of the GM-PHD filter, which can be able to correctly redistribute the weights of closely spaced targets, and effectively improve the multiple target state estimation precision. Simulation results demonstrate that the proposed algorithm can accurately estimate the number of targets and their states, and effectively improve the performance of multi-target tracking algorithm.

scholarly journals Strong Tracking PHD Filter Based on Variational Bayesian with Inaccurate Process and Measurement Noise Covariance

Sensors ◽  
2021 ◽  
Vol 21 (4) ◽  
pp. 1126
Author(s):  
Zhentao Hu ◽  
Linlin Yang ◽  
Yong Jin ◽  
Han Wang ◽  
Shibo Yang
Keyword(s):  
Real Time ◽  
Gaussian Mixture ◽  
Measurement Noise ◽  
Tracking Accuracy ◽  
Process Noise ◽  
Probability Hypothesis Density ◽  
Measurement Noise Covariance ◽  
Multi Target Tracking ◽  
Noise Covariance ◽  
Fading Factor

Assuming that the measurement and process noise covariances are known, the probability hypothesis density (PHD) filter is effective in real-time multi-target tracking; however, noise covariance is often unknown and time-varying for an actual scene. To solve this problem, a strong tracking PHD filter based on Variational Bayes (VB) approximation is proposed in this paper. The measurement noise covariance is described in the linear system by the inverse Wishart (IW) distribution. Then, the fading factor in the strong tracking principle uses the optimal measurement noise covariance at the previous moment to control the state prediction covariance in real-time. The Gaussian IW (GIW) joint distribution adopts the VB approximation to jointly return the measurement noise covariance and the target state covariance. The simulation results show that, compared with the traditional Gaussian mixture PHD (GM-PHD) and the VB-adaptive PHD, the proposed algorithm has higher tracking accuracy and stronger robustness in a more reasonable calculation time.

scholarly journals Sensor Selection for Decentralized Large-Scale Multi-Target Tracking Network

Sensors ◽  
2018 ◽  
Vol 18 (12) ◽  
pp. 4115 ◽  
Author(s):  
Feng Lian ◽  
Liming Hou ◽  
Bo Wei ◽  
Chongzhao Han
Keyword(s):  
Target Tracking ◽  
Large Scale ◽  
Sequential Monte Carlo ◽  
Signal To Noise Ratio ◽  
Computational Cost ◽  
Gaussian Mixture ◽  
Descent Method ◽  
Sensor Selection ◽  
Multi Target Tracking ◽  
Finite Set

A new optimization algorithm of sensor selection is proposed in this paper for decentralized large-scale multi-target tracking (MTT) network within a labeled random finite set (RFS) framework. The method is performed based on a marginalized δ-generalized labeled multi-Bernoulli RFS. The rule of weighted Kullback-Leibler average (KLA) is used to fuse local multi-target densities. A new metric, named as the label assignment (LA) metric, is proposed to measure the distance for two labeled sets. The lower bound of LA metric based mean square error between the labeled multi-target state set and its estimate is taken as the optimized objective function of sensor selection. The proposed bound is obtained by the information inequality to RFS measurement. Then, we present the sequential Monte Carlo and Gaussian mixture implementations for the bound. Another advantage of the bound is that it provides a basis for setting the weights of KLA. The coordinate descent method is proposed to compromise the computational cost of sensor selection and the accuracy of MTT. Simulations verify the effectiveness of our method under different signal-to- noise ratio scenarios.

Centralized multi-sensor multi-target tracking with labeled random finite set

2016 ◽  
Vol 231 (4) ◽  
pp. 669-676 ◽  
Author(s):  
Hao Qiu ◽  
Gaoming Huang ◽  
Jun Gao
Keyword(s):  
Probability Hypothesis Density ◽  
Multiple Sensors ◽  
Parallel Track ◽  
Truncation Scheme ◽  
Moment Approximation ◽  
Multi Target Tracking ◽  
Finite Set ◽  
Density Filter ◽  
Single Sensor ◽  
First Moment

Tracking multiple objects with multiple sensors is widely recognized to be much more complex than the single-sensor scenario. This contribution proposes a computationally tractable multi-sensor multi-target tracker. Based on Bayes equation and multi-senor observation model, a new corrector for multi-senor is derived. To lower the complexity of update operation, a parallel track-to-measurement association strategy is applied to the corrector. Hypotheses truncation scheme along with first-moment approximation of multi-target density are also employed to improve the tracking efficiency. The tracker is applied to a couple-sensor scenario. Experiment results validate the advantages of proposed method compared to the standard single-sensor δ-generalized labeled multi-Bernoulli filter and the iterated-corrector probability hypothesis density filter.

Convolution PHD Filtering for Nonlinear Non-Gaussian Models

2011 ◽  
Vol 213 ◽  
pp. 344-348
Author(s):  
Jian Jun Yin ◽  
Jian Qiu Zhang
Keyword(s):  
Parallel Implementation ◽  
Gaussian Mixture ◽  
Probability Hypothesis Density ◽  
Gaussian Sum ◽  
Multi Target Tracking ◽  
Lower Complexity ◽  
Observation Noise ◽  
Non Gaussian ◽  
Convolution Filters ◽  
Cphd Filter

A novel probability hypothesis density (PHD) filter, called the Gaussian mixture convolution PHD (GMCPHD) filter was proposed. The PHD within the filter is approximated by a Gaussian sum, as in the Gaussian mixture PHD (GMPHD) filter, but the model may be non-Gaussian and nonlinear. This is implemented by a bank of convolution filters with Gaussian approximations to the predicted and posterior densities. The analysis results show the lower complexity, more amenable for parallel implementation of the GMCPHD filter than the convolution PHD (CPHD) filter and the ability to deal with complex observation model, small observation noise and non-Gaussian noise of the proposed filter over the existing Gaussian mixture particle PHD (GMPPHD) filter. The multi-target tracking simulation results verify the effectiveness of the proposed method.

Sign in / Sign up

Export Citation Format

Share Document